Research on Alzheimer?s disease (AD) has strongly focused on pathological alterations detectable by histopathology or radiological imaging such as plaques and tangles. However, drug treatments targeting these alterations have not slowed cognitive decline in AD patients despite clear evidence for target engagement in the brain. Efforts to discover biomarkers for AD have also focused primarily on neuropathological alterations. To pursue novel directions and fill important knowledge gaps in the field, this proposal focuses on functionally relevant neural network abnormalities that likely underlie cognitive deficits and could promote neurodegeneration and AD progression through diverse mechanisms, including dysfunction of microglia, the innate immune cells of the brain. Previously, we found subclinical epileptic activity in patients with AD and in related mouse models. In the patients, the extent of such network abnormalities predicted cognitive decline. In the mouse models, prevention or reversal of the network dysfunction improved survival and reduced cognitive deficits. More recently obtained preliminary data suggest a novel pathogenic link between network and immune cell dysfunction. Suppressing epileptic activity reduced microglial activation in mice with elevated amyloid-b (Ab) levels in the brain. Knock-in mice with microglial dysfunction, induced by a human immune gene variant that increases AD risk, had increased and prolonged epileptic activity after challenge with an excitotoxin. Nonconvulsive epileptic activity in mice with Ab accumulation in the brain correlated with levels of specific inflammatory mediators in plasma. Based on these intriguing findings, we propose to test the novel and unifying hypothesis that aberrant neural network activity and immune dysfunction engage in a vicious cycle that promotes synaptic loss, the likeliest cause of cognitive decline in AD. Conceivably, either of these components can initiate the cycle, possibly at different times in different patients and in response to diverse triggers, including injury-induced surges in amyloid-b (Ab) or tau levels and aging-related alterations. Which of the components is triggered first may depend on a person?s genetics, including genes affecting microglial functions. We will use AD-related mouse models to begin to test these hypotheses at the preclinical level. Specifically, we will determine whether (1) suppression of neural network abnormalities reduces aberrant microglial activation and preserves synaptic density, (2) network dysfunction and synaptic deficits are reflected by immune markers in peripheral blood, and (3) microglial dysfunction contributes to aberrant network activity and synaptic loss. The proposed studies will help delineate the role of immune cells in AD-related network dysfunction and synaptic impairment. They may also identify potential new biomarkers, such as EEG signatures and peripheral blood alterations, and could help identify therapeutic strategies to modulate immune cell activities that may ultimately benefit patients with AD and people at risk of developing this disorder.